Opportunistic screening for incident cardiometabolic disease in metabolically healthy non-obese individuals: a prospective cohort study

Figures

• 

  Fig. 1. Study Overview. a Deep learning framework utilizing 3D U-Net for whole-body MRI segmentation to quantify volumetric SAT and VAT. b Two cohorts were analyzed: NAKO (N = 23,664) for cutoff derivation and UKB (N = 36,311) with longitudinal follow-up data. (c) Study workflow: Sex-specific VAT/SAT ratio cutoffs were derived from NAKO using ROC analysis based on prevalent cardiometabolic disease, then applied to UKB MHN individuals (N = 22,040; subset of the full UKB cohort after applying metabolic health and obesity criteria) to assess reclassification performance and prognostic value for incident MACE and diabetes. BMI, body mass index; F, female; M, male; MACE, major adverse cardiovascular events; MHN, metabolically healthy non-obese; MRI, magnetic resonance imaging; NAKO, German National Cohort; ROC, receiver operating characteristic; SAT, subcutaneous adipose tissue; UKB, UK Biobank; VAT, visceral adipose tissue

• 

  Fig. 2. Consort diagram. Flowchart demonstrating participant selection and exclusion criteria for the UKB cohort. 3D BC, three-dimensional body composition; BMI, body mass index; MACE, major adverse cardiovascular events; MRI, magnetic resonance imaging; MHN, metabolically healthy non-obese; NAKO, German National Cohort; UKB, UK Biobank; SAT subcutaneous adipose tissue; VAT, visceral adipose tissue; WC, waist circumference

• 

  Fig. 3. Distribution of VAT/SAT ratio and waist circumference, and reclassification analysis. a Scatter plots showing the distribution of participants by sex across four risk groups defined by sex-specific cutoffs for WC (80 cm for females, 94 cm for males, red and blue vertical dashed lines) and VAT/SAT ratio (0.15 for females, 0.35 for males, red and blue horizontal dashed lines). Each point represents one participant. b Risk stratification matrices displaying event rates for incident MACE and diabetes across the four risk groups. Numbers indicate events, total participants, and event rates (%) in each cell. Color intensity reflects event rate magnitude (white to dark blue gradient). MACE, major adverse cardiovascular events; SAT, subcutaneous adipose tissue; VAT, visceral adipose tissue; WC, waist circumference

• 

  Fig. 4. Associations of VAT/SAT ratio with incident MACE and diabetes in metabolically healthy non-obese individuals. a − b Cumulative incidence curves for MACE and diabetes stratified by VAT/SAT ratio groups (high vs. low based on sex-specific cutoffs) with numbers at risk shown below each plot. Forest plots display hazard ratios from Cox proportional hazards models with stepwise adjustment. Model 1: unadjusted; Model 2: adjusted for age, sex, and smoking status; Model 3: Model 2 + waist circumference; Model 4: Model 2 + BMI; Model 5: Model 2 + waist circumference + BMI. Log-rank p < 0.001 for both outcomes. BMI, body mass index; HR, hazard ratio; MACE, major adverse cardiovascular events; SAT, subcutaneous adipose tissue; VAT, visceral adipose tissue